Physiotherapy informed by Acceptanceand Commitment Therapy (PACT):protocol for a randomised controlledtrial of PACT versus usualphysiotherapy care for adultswith chronic low back pain

Emma Godfrey,1 Melissa Galea Holmes,1 Vari Wileman,1 Lance McCracken,1

Sam Norton,1 Rona Moss-Morris,1 John Pallet,2 Duncan Sanders,3

Massimo Barcellona,4 Duncan Critchley5

To cite: Godfrey E, GaleaHolmes M, Wileman V, et al.Physiotherapy informed byAcceptance and CommitmentTherapy (PACT): protocol fora randomised controlled trialof PACT versus usualphysiotherapy care for adultswith chronic low back pain.BMJ Open 2016;6:e011548.doi:10.1136/bmjopen-2016-011548

▸ Prepublication history forthis paper is available online.To view these files pleasevisit the journal online(http://dx.doi.org/10.1136/bmjopen-2016-011548).

Received 17 February 2016Revised 25 April 2016Accepted 29 April 2016

For numbered affiliations seeend of article.

Correspondence toDr Emma Godfrey;emma.l.godfrey@kcl.ac.uk

ABSTRACTIntroduction: Chronic low back pain (CLBP) is acommon condition and source of significant suffering,disability and healthcare costs. Current physiotherapytreatment is moderately effective. Combining theory-based psychological methods with physiotherapy couldimprove outcomes for people with CLBP. The primaryaim of this randomised controlled trial (RCT) is toevaluate the efficacy of Physiotherapy informed byAcceptance and Commitment Therapy (PACT) onfunctioning in patients with CLBP.Methods and analysis: The PACT trial is a two-armed, parallel-group, multicentre RCT to assess theefficacy of PACT in comparison with usualphysiotherapy care (UC). 240 patients referred tophysiotherapy with CLBP will be recruited from threeNational Health Service (NHS) hospitals trusts.Inclusion criteria are: age ≥18 years, CLBP ≥12-weekduration, scoring ≥3 points on the Roland-MorrisDisability Questionnaire (RMDQ) and adequateunderstanding of spoken and written English toparticipate. Patients will be randomised to PACT or UC(120 per arm stratified by centre) by an independentrandomisation service and followed up at 3 and12 months post randomisation. The sample size of 240will provide adequate power to detect a standardisedmean difference of 0.40 in the primary outcome(RMDQ; 5% significance, 80% power) assumingattrition of 20%. Analysis will be by intention to treatconducted by the trial statistician, blind to treatmentgroup, following a prespecified analysis plan.Estimates of treatment effect at the follow-upassessments will use an intention-to-treatframework, implemented using a linear mixed-effectsmodel.Ethics and dissemination: This trial has full ethicalapproval (14/SC/0277). It will be disseminated viapeer-reviewed publications and conferencepresentations. The results will enable clinicians,patients and health service managers to make informed

decisions regarding the efficacy of PACT for patientswith CLBP.Trial registration number: ISRCTN95392287;Pre-results.

INTRODUCTIONLow back pain has a lifetime prevalenceranging from 60% to 70% in industrialisedcountries, causes more years of disabilitythan any other health condition and is thesecond most frequent reason for absencefrom work.1 2 Chronic low back pain (CLBP)

Strengths and limitations of this study

▪ The Physiotherapy informed by Acceptance andCommitment Therapy (PACT) trial will be the firstrandomised controlled trial to test the efficacy ofa physiotherapist-led ACT-informed interventionfor chronic low back pain (CLBP) against stand-ard physiotherapy.

▪ The PACT trial will assess the feasibility of train-ing physiotherapists to deliver a novel psycho-logically informed physiotherapy intervention.

▪ Theory-based processes of change consistentwith the psychological flexibility model will beevaluated, providing evidence for the mechani-sms underpinning observed outcomes.

▪ Restriction to participants referred to physiother-apy services and speaking English may limit gen-eralisability of findings.

▪ Patients who have had prior treatment frommultidisciplinary or cognitive–behaviouraltherapy (CBT) pain management at any time andother physiotherapy treatment in the previous6 months will be excluded due to possible con-tamination effects.

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 1

Open Access Protocol

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

is pain that has lasted for more than 12 weeks. It causesconsiderable suffering to the individual and is a majorfinancial burden on the National Health Service (NHS)and wider society. UK healthcare costs are £1.6 billionannually,3 and CLBP is responsible for 80% of this cost.4

Physiotherapy is a common treatment for CLBP, with1.26 million patients referred to NHS physiotherapists ata cost of £150 million per annum.5 Several forms ofphysiotherapy are recommended for CLBP, includingexercises, manual therapy and back classes.6 The type ofphysiotherapy delivered varies considerably in durationand content, and there is little consensus about themost appropriate and cost-effective treatment.7 8 Manytrials show no clear superiority for any treatment, withthe majority leading to no more than modest improve-ment in pain and disability outcomes.9 As a result,patients are often overtreated, placing high demands onphysiotherapy services and delaying active self-management. This highlights the need to develop andtest novel treatments for patients with CLBP.10

CLBP is best suited to a biopsychosocial model ofcare11 and a cognitive behavioural approach to treat-ment.12 Cognitive–behavioural therapy (CBT) has agood evidence base for the treatment of chronicpain.13–15 A Cochrane review concluded that furthergeneral randomised controlled trials (RCTs) of CBT forchronic pain were not required.16 Instead, studies identi-fying the specific components of CBT and attempting tounderstand which underlying processes were successfulwere recommended. The Chartered Society ofPhysiotherapy recognises that CBT can fall within a phy-siotherapist’s scope of practice.17 However, CBT-basedtreatments delivered by physiotherapists have onlyproduced moderate improvements in CLBP-relateddisability,18 19 and many physiotherapists do not feeladequately trained to use psychological techniqueseffectively.20 There is potential for enhancing effective-ness through greater focus on competency, but itremains unclear how to best implement cognitive andbehavioural approaches during physiotherapyinterventions.One promising theory-based approach to chronic pain

is a form of CBT called Acceptance and CommitmentTherapy (ACT).21 22 ACT has been shown to have posi-tive effects in chronic pain,23 24 and meta-analyses ofACT for chronic pain showed improvements in depres-sion, anxiety, pain intensity, physical functioning andquality of life (QoL).25 26 ACT aims to increase psycho-logical flexibility and focuses on improving functionrather than reducing pain. It has good maintenance oftreatment effects up to 3 years post treatment,27 import-ant in a chronic relapsing and remitting condition suchas CLBP. In all published studies to date, ACT has beendelivered by psychologists or within multidisciplinaryteams; however, psychology is a limited resource andmost patients with CLBP are seen by physiotherapists.A recent trial of ACT for CLBP delivered by psycholo-gists found that patients referred for physiotherapy were

somewhat resistant to seeing a psychologist and conse-quently has recommended combining ACT with physio-therapy.28 A qualitative study investigated potentialbarriers and facilitators to embedding ACT within aphysiotherapist-led pain rehabilitation programme.Findings suggested this presented challenges and oppor-tunities but was a positive experience overall if extrasupport was provided.29

We have developed a brief physiotherapist-deliveredtreatment, guided by principles of ACT, Physiotherapyinformed by Acceptance and Commitment Therapy(PACT), consisting of two face-to-face sessions plus afollow-up telephone call. A small proof-of-concept feasi-bility study demonstrated the acceptability of the inter-vention for patients, and recruitment to a larger trialwas achievable.30 This protocol describes a phase II effi-cacy RCT using a two-armed, parallel-group design toassess the efficacy of PACT for improving function at3 months in individuals with CLBP, in comparison withusual physiotherapy treatment. Across three NHS trusts(including six hospital centres), 240 people with CLBPwill be individually randomised to PACT or usual physio-therapy care (UC). We hypothesise that the groupreceiving PACT will have improved self-reported func-tioning at the primary end point of 3-month follow-upcompared with UC. The PACT trial is funded by theNational Institute for Health Research (NIHR) Researchfor Patient Benefit programme, reference numberPB-PG-1112-29055.

METHODS AND ANALYSISMain research questionWhat is the efficacy of PACT for improving functioningin patients with CLBP?

Research objectivesPrimary objectives1. The primary objective of this study is to evaluate the

efficacy of PACT on the primary end point of func-tioning at 3-month follow-up.

Secondary objectives1. To assess whether PACT has a positive impact on sec-

ondary outcomes: QoL and function in variousdomains, process variables such as acceptance andcommitted action, mood, self-efficacy and pain com-pared with UC at 3-month and 12-month follow-up.

2. To investigate optimal ways of training physiothera-pists to work in extended roles and develop a PACTtraining package for use in a definitive multicentretrial.

3. To pilot methods and instruments needed to estimatecost-effectiveness in a future phase III trial from ahealth service and societal perspective.

4. To assess the acceptability of the intervention andtraining for patients and clinicians via nested qualita-tive studies.

2 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

5. To investigate hypothesised processes of clinicalimprovement following PACT, including predictorsand moderators of outcome, and treatment fidelity.

DesignA phase II, assessor-blind, multicentre, two-armed,parallel-group RCT.

MethodA total of 240 patients with CLBP will be individuallyrandomised to PACT or UC.

SettingParticipants will be recruited from secondary care physio-therapy clinics in two NHS Foundation Hospital trusts inLondon (Guy’s and St Thomas’ and King’s CollegeHospital) and one in the South East of England (Ashfordand St Peter’s), UK. Treatment will take place in thephysiotherapy clinics based at the participating hospitals.

EligibilityInclusion criteria: Adults (aged 18 years and over) withnon-specific CLBP (confirmed by a clinical physiother-apist) of >12 weeks duration, with or without associatedleg pain and reporting a score of 3 points or more onthe Roland-Morris Disability Questionnaire (RMDQ).Patients need to be able and willing to provide informedconsent and attend treatment at hospital. Potential parti-cipants require a good understanding of spoken andwritten English to complete trial data collection and par-ticipate in the PACT programme.Exclusion criteria: Prior treatment from multidisciplin-

ary CBT pain management at any time and otherphysiotherapy treatment in the previous 6 months orinjection therapy within 3 months. Specific medicallydiagnosed lumbar spine pathology (eg, inflammatoryarthritis, fracture or cancer). Patients with deterioratingneurological signs (stable neurological signs and pain ofapparently neuropathic origin are not exclusion criteria)and those with previous experience of or awaiting spinalsurgery. Patients with current psychiatric illness (eg,severe depression, personality disorder or post-traumaticstress disorder) and/or current drug or alcohol misuselikely to interfere with treatment.Withdrawal criteria: Participants will be withdrawn from

the trial if there are any concerns regarding informedconsent. Participants can also withdraw if they choose towithout giving a reason. If patients withdraw consent forresearch follow-up during the trial, reasons for dropoutwill be recorded where possible.

Planned interventionsPhysiotherapy informed by Acceptance and CommitmentTherapyPACT is a brief physiotherapy intervention guided byprinciples of ACT designed to promote self-management,consisting of two 60 min face-to-face sessions 2 weeksapart, plus one booster telephone call (lasting 20 min),

1 month after the last treatment session. PACT alters thecontent of physiotherapy treatment and reconfigures itso that it is delivered in fewer but longer sessions,although the total contact time is similar to the averageamount of time patients with CLBP receive as part ofusual physiotherapy treatment as reported in two UKRCTs for CLBP, where usual physiotherapy was used asthe control arm.31 32 Two 60-min sessions are designedto allow adequate time to do an initial physical assess-ment and provide feedback, create value-based goals,provide individualised physical exercises and teachsimple psychological skills to promote psychologicalflexibility, and to address facilitators and barriers to self-management. The booster phone call promotes self-management by giving patients a chance to feedbackprogress and gain support with any ongoing issues theymay have. PACT thus aims to directly reduce avoidanceand promote openness, to build present-focused aware-ness, and coordinate greater engagement in goal-oriented and value-based activity (see box 1 below). Theface-to-face intervention will be supported by a patientmanual individualised to patient needs. Patients rando-mised to PACT will be given their patient manual duringtheir first session.

Training and supervisionPACT will be delivered by 8 Band 6 or 7 trial phy-siotherapists (2 per centre). Physiotherapists will beidentified by their managers and invited to volunteer totake part in the study. The physiotherapist will then besent information about the PACT study and be invited tomeet the study team to discuss their participation.Training will be provided by LM, a clinical psychologistand expert in ACT, with the assistance of EG, a healthpsychologist, and DC, a physiotherapist, before the startof recruitment. Group face-to-face training includingexperiential exercises and role play will last 2 days andwill be supported by a manual. The manual consists ofan introduction to ACT and promoting behaviourchange; information about the trial; strategies, meta-phors and skills to enable PACT delivery; detailedsession plans (see box 1); explanation of competencyand fidelity, including the use of supervision and areflexive diary. Obstacles to therapist and patient engage-ment and progress will be discussed, as well as strategiesfor dealing with these eventualities. The trial protocolwill be reviewed, including recording the timing andlength of sessions, any deviations from protocol includ-ing missed sessions or dropout, and confidential storageof audio recordings. A training package will be furtherdeveloped through feedback forms filled in at the endof the group face-to-face training and via interviews withall trained physiotherapists as part of this study, toenable its use in a larger phase III trial if PACT is suc-cessful. Each physiotherapist will practise deliveringPACT and receive at least two sessions of individualsupervision to ensure adequate competency to com-mence treatment. As PACT is a novel treatment, we will

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 3

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

assess competency qualitatively through the training andinitial supervision process, which will include listening toaudiotaped sessions and observing role play.Physiotherapists will also be asked to report back onexperiences with practice patients before they start thetrial. If a physiotherapist is not deemed competent tobegin delivery after two individual sessions of supervi-sion, they will be offered more sessions until a satisfac-tory level of competency is observed. We will continue toassess competency throughout the trial on a monthlybasis. It is assumed competency will improve during thecourse of delivery as skills are enhanced through prac-tice and supervision. Trial physiotherapists will attendmonthly supervision meetings with supervisors (LM, EGand DC), to maintain skills and receive support. Regularsupervision will ensure that the physiotherapists adhereto the trial protocol and that the quality of the interven-tion is maintained. Fidelity to the treatment protocolwill also be enhanced by the use of session checklistsand ratings of audio tapes from the trial with feedbackprovided to clinicians.

Treatment fidelityAll PACT sessions will be audio-recorded for the purposeof assessing treatment fidelity. These will be used forsupervision during the study and to check fidelitythroughout the trial. Supervisors will listen to one tapeper physiotherapist per month. Once the trial hasended, a subset of the audio recordings will be analysedfor overall fidelity. These fidelity checks will be under-taken via assessment of a sample of audio recordings ofPACT sessions, across sites and physiotherapists, under-taken by two independent researchers. A modified fidel-ity measure will be developed based on the Plumb andVilardaga33 paper and the ACT for Chronic PainAdherence Rating Scale used in the OptimisedBehavioural Intervention trial.28 At least two sessionsfrom every physiotherapist will be rated in terms ofadherence to the manual and checklist. The therapeuticalliance between physiotherapists and participants willalso be rated using a therapy process scale34 employedin previous RCTs of treatments for chronic fatigue and aweight loss intervention in primary care.

Usual careParticipants randomised to UC will receive any treat-ment considered suitable by their treating physiother-apist. Treatment may include any type of individualphysiotherapy and/or back classes, for example, exer-cises, manual therapy, hydrotherapy and back schools.Consistent with the Consolidated Standards ofReporting Trials (CONSORT) guidelines for complexinterventions,35 we will collect data on volume (durationand frequency of sessions) and components (eg,one-to-one treatment vs group exercise class) of treat-ment received by participants in the UC arm, and weplan to report and publish these essential details of thecontrol condition with trial results.

Box 1 Summary of the content of Physiotherapyinformed by Acceptance and Commitment Therapy (PACT)sessions

PACT Session 1: 60 min face to face▸ Set the agenda: outline structure, schedule and delivery of

treatment.▸ Assessment, feedback and rationale: conduct brief physical

assessment and discuss results. Empathise with and normal-ise current feelings and provide guidance that no seriousmedical problems have been uncovered and it is safe toresume normal activities.

▸ Shifting focus from pain to function: discuss previousattempts to reduce pain, which are not usually very successfulin relation to daily functioning. Build open engagement ratherthan struggling with pain. Present the goal of PACT, to helppeople function better, especially in the areas that are import-ant to them. Use metaphors to help make this shift.

▸ Value-based goal setting: introduce patient manual. Engagepatient in identifying core values and setting related goals.Break goals down into small steps that are positive, practicaland achievable, and record these in the manual.

▸ Skills training to address barriers to goal attainment: imple-ment strategies to promote openness, awareness and engage-ment, for example, mindfulness exercises, action plans andmaking a public commitment to goals, to help anticipate andovercome perceived barriers to change.

PACT Session 2: 60 min face to face▸ Review successes and challenges: positively reinforce pro-

gress towards goals, discuss how this was achieved and high-light benefits. Review, normalise and empathise withchallenges and encourage continued use of the patientmanual.

▸ Goal adjustment/development: check the salience of goals andmake adjustments if required. Re-establish commitment usingmotivational interviewing techniques if necessary. Use exer-cises and metaphors to normalise setbacks, keep moving insmall steps toward goals and troubleshoot or prevent theeffects of barriers.

▸ Generalisation to new areas: rehearse new skills, such asmindfulness and shifting focus and explore how these can beextended to other areas of life. Encourage the development ofinsights and the capacity to self-initiate change.

▸ Integration of self-management approach: review key skillsand identify a support network. Discuss maintenance toolsand again normalise setbacks.

PACT Booster Call: 20-min phone call▸ Review progress: appreciate successes to date and discuss

any remaining barriers.▸ Assessment of skills integration into everyday life: review key

skill sets so that they organise the participant’s learning in theareas of openness, awareness and engagement.

▸ Support generalisation: build on patterns of initial goalachievement and broaden the scope of applications to otherareas.

▸ Reinforce continued self-management: emphasise to thepatient that they will face times in the future when they experi-ence pain or other difficulties and they have resources to dealwith this, such as the patient manual and new skills. Positiveclosure of the therapeutic partnership to help reinforce theircapacity to persist with the tools they have to manage theirback pain without needing more healthcare.

4 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

Separate groups of clinicians will deliver PACT and UCto avoid contamination. We will explicitly inform PACTphysiotherapists about the risks and consequences ofcontamination during training and supervision and willask that they do not share material or ideas with theircolleagues during the treatment delivery period. In addi-tion, we will ensure that all PACT sessions are conductedin private rooms to eliminate the possibility of UC phy-siotherapists overhearing what is being provided in thenovel treatment arm.

Participant identification and recruitmentIn total, 240 patients will be recruited, 120 per treatmentarm, over an 18-month period. Patients will be recruitedfrom six secondary care physiotherapy clinics in Londonand the South East of England. Posters advertising thestudy will be placed in relevant physiotherapy clinics inorder to inform patients and clinicians about the study.Potential participants referred to outpatient physiother-apy by their general practitioner (GP) or consultant willbe identified by clinical physiotherapists from each hos-pital centre at their initial triage sessions, provided withwritten and verbal information about the PACT trial,and invited to participate. Participants who consent tobe contacted will be referred to the research associates(RAs) for full eligibility screening, conducted by tele-phone. All patients who undergo screening will berecorded anonymously on a screening database asso-ciated with the study by the RAs. If the patient is suit-able, the RA will then invite them to complete consentforms and baseline measures either at home online, viapostal questionnaires or in person at the clinic (table 1).GPs will be notified in writing of their patient’s partici-pation. Patients will be informed that they can withdrawfrom the study at any time without giving a reason andthat this will not affect the treatment they receive in anyway. All participant responses will be anonymous andconfidential, and participants will not be identified inany way by their responses (figure 1).

Study proceduresInformation on study procedures is summarised in theCONSORT diagram (figure 1) and table 1 (screeningand data collection).

RandomisationRandomisation will be provided by an independent ran-domisation service at the UK Clinical ResearchCollaboration (UKCRC)-registered King’s Clinical TrialsUnit (CTU). Randomisation will be at the level of theindividual, using block randomisation with randomlyvarying block sizes, stratified by centre and implementedvia the King’s CTU online system, with emails generatedautomatically and sent to relevant physiotherapy staff atstudy sites.

BlindingIt is not possible to blind patients or treating phy-siotherapists to the treatment allocation; however, nohypotheses have been proposed to participants as tothe superiority of PACT over UC and all participantswill receive physiotherapy treatment. The patient infor-mation sheet will deliberately maintain a position ofequipoise by stating, ‘Each group will get a treatmentthat we think might be helpful, but we don’t knowwhether one treatment is going to be more helpfulthan another’.The RAs screening patients and collecting data and

the statistician analysing the data and assessing outcomewill be blinded to treatment allocation. All outcomes arepatient reported and collected via the internet followingautomated email reminders, reducing the risk ofunblinding the assessors. Locked codes will be used fortreatment allocation, and the trial statistician will analysethe data blind.

Data collectionParticipant screening data will be collected by telephoneand entered into the database by the RAs. Baseline,3-month and 12-month follow-up data will be collectedthrough self-report questionnaires. The RAs, blind totreatment allocation, will administer questionnaires andconduct data entry. Treatment allocation will not beincluded on the questionnaires. Research data will beentered onto the MedSciNet database system, a regula-tory compliant database that has been enabled foronline collection of patient-reported outcome measures(PROMS). Participants will be given a unique usernameand password to log into the online database and com-plete consent and measures at each time point. Theirdata will be identified by a unique identification numberand will be kept separate from any personal identifyingdata to maintain confidentiality. Baseline and outcomedata will be patient self-completed at home (eitheronline or via postal questionnaires), thus avoiding anyinfluence of the study team on the responses and redu-cing bias. PACT physiotherapists will have access to aunique database on the MedSciNet system to recorddetails of who provided PACT and UC sessions, thenumber of sessions attended and any dropouts, as wellas the number and type of UC treatment sessionsattended.

Outcome measuresTime points: Assessments will be completed at baseline(immediately pre-randomisation), and 3 months and12 months post randomisation by all participants. Alltime points will be taken into account during analysis,but the primary efficacy end point is 3-month follow-up.In order to justify treatment costs, clinically significanttreatment effects need to be maintained over time, andthis is particularly important in a chronic relapsing andremitting condition such as CLBP, so maintenance ofany treatment effects will be assessed at 12 months. The

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 5

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

RAs will be employed to coordinate the trial and collectbaseline and follow-up data. All participants will be sent(emailed and posted) follow-up questionnaires by theRAs at 3 and 12 months. Participants not returning ques-tionnaires within 1 week will receive a reminder email,telephone call and text in 3-day intervals. One weekafter that, if no data have been returned, the RAs will

ring the participant to ask if they can collect primaryoutcome data over the telephone.

Baseline measuresParticipants will complete a baseline assessment question-naire which includes the validated scales detailed below,plus demographic data to establish sociodemographic

Table 1 Screening and data collection across the trial: summary of the key trial processes from a potential participant

agreeing to be contacted to the data collection time points

Process

Completed

by

Format of

administration Pre-consent Baseline

3-Month

follow-up

12-Month

follow-up

Ongoing

during

treatment

period Reference

Identification* PT PP ●Screening RA Telephone ●Consent P PP/DB ●Randomisation CTU CTU Database ●Sociodemographics P PP/DB ●EuroQol-5D-5L P PP/DB ● ● ● 36

Medical Outcomes

Survey Short

Form-12 (V.2)

P PP/DB ● ● ● 37

Roland-Morris

Disability

Questionnaire

P PP/DB ● ● ● 38

Chronic Pain

Acceptance

Questionnaire-8

P PP/DB ● ● ● 39

Committed Action

Questionnaire-8

P PP/DB ● ● ● 40

Numeric Analogue

Scale

P PP/DB ● ● ● –

Patient-Specific

Function Scale

P PP/DB ● ● ● 41

Work and Social

Adjustment Scale

P PP/DB ● ● ● 42

Pain Self-Efficacy

Questionnaire

P PP/DB ● ● ● 43

Generalised Anxiety

Disorder-7

P PP/DB ● ● ● 45

Life Satisfaction

Scale

P PP/DB ● ● ● –

Patient Health

Questionnaire-9

P PP/DB ● ● ● 46

Global improvement P PP/DB ● ● 47

Satisfaction with

outcome

P PP/DB ● ● 47

Treatment credibility P PP/DB ● ● 48

Health-related

resource use

P PP/DB ● ● ● 49

Self-reported

adverse event

P PP/DB ● ● –

Clinician-reported

adverse event

PT PP/DB ● –

Treatment

attendance

PT DB ● –

*Includes permission to contact and provision of patient information letter.CTU, King’s Clinical Trials Unit; DB, online database; P, participant; PP, paper and pencil; PT, physiotherapist; RA, research associate.

6 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

characteristics of participants as follows: age, sex, height,weight, self-reported ethnicity, education level, employ-ment and benefit status, diagnosis and history of anymedical condition if available.

Primary outcome: the RMDQPatient-reported disability is recommended as a coreoutcome measure in low back pain14 and chronic paintrials.50 The Roland-Morris Disability Questionnaire(RMDQ)38 is a 24-item questionnaire assessing self-reported functioning and disability due to CLBP,ranging from 0 (no disability) to 24 (maximum disabil-ity). The RMDQ is a widely used and valid measure withgood test–retest reliability. A 2–3 point change frombaseline is considered clinically important.51

Secondary outcome measuresSecondary outcomes have been selected to determinethe wider effects of PACT and to assess therapeutic pro-cesses and mechanisms of action. The outcomes includeall core domains recommended in chronic painresearch:50 pain, function, mood, QoL and satisfactionwith treatment. A Global Improvement scale47 and

Treatment Credibility48 Questionnaire will be completedat follow-up.

QoL: Work and Social Adjustment Scale (WSAS) andEQ-5D-5LThe Work and Social Adjustment Scale (WSAS) measuresthe effect of CLBP on participants’ ability to work and par-ticipate in social and private leisure activities.42 The WSAShas five items scored from 0 (not affected) to 8 (severelyaffected), with a total possible score of 40. The EQ-5D-5L isthe most frequently used tool for generating quality-adjustedlife years (QALYs), which are favoured by the NationalInstitute for Health and Care Excellence (NICE).52

Pain: pain Numeric Analogue Scale (NAS)A single pain item rated using a numerical analoguescale anchored at 0 with ‘no pain’ and 10 with ‘worstpossible pain’ will reflect the participants’ subjectiveexperience of pain.

Function: Patient-Specific Functional Scale (PSFS)The PSFS is a self-reported measure used to identify andinvestigate functional status tailored to the patient.41

Figure 1 PACT trial process

flow chart. CTU, King’s Clinical

Trials Unit; PACT, Physiotherapy

informed by Acceptance and

Commitment Therapy. aGuy’s and

St Thomas’ Hospitals, King’s

College Hospital and Ashford and

St Peters Hospitals.

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 7

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

The patient identifies three activities limited by theirCLBP, rating them on a scale of 0 (unable to performactivity) to 10 (able to perform activity at the same levelas before injury/problem). The scores across the threeitems are summed to give a total possible score of 30.

Mood: Generalised Anxiety Disorder-7 (GAD-7) and PatientHealth Questionnaire-9 (PHQ-9)The GAD-745 has seven items that assess anxiety in thelast 2 weeks. Scores range from 0 to 21, with a total scoreof ≥8 indicating probable generalised anxiety disorder.The PHQ-946 is a brief nine-item questionnaire thatidentifies and quantifies depressive symptoms; scoresrange from 0 to 27, with a total score of ≥10 indicatingprobable depressive disorder. Both questionnaires arewell validated, commonly used self-report instrumentsfor detecting distress, depression and anxiety in patientswith medical illnesses.

Process variablesThe Chronic Pain Acceptance Questionnaire-8 (CPAQ-8) andCommitted Action Questionnaire-8 (CAQ-8)Acceptance of pain and persistent but flexible behaviourtowards achieving a goal form part of the ACT modeland are therefore putative mediators of the efficacymechanism in PACT treatment. The CPAQ-839 is a shor-tened version of the original 20-item Chronic PainAcceptance Questionnaire, which assesses the capacityto engage in activities without struggling with the pain.Each item is scored from 0 (never true) to 6 (alwaystrue), with a total possible score of 48. The CAQ-840 is ashortened version composed of eight questions from theoriginal 18-item Committed Action Questionnaire aimedto measure committed action in terms of commitmentto valued goals. The items are rated from 0 (never true)to 6 (always true), with a total possible score of 48.

Pain Self-Efficacy Questionnaire (PSEQ)The PSEQ43 assesses confidence in undertaking normalactivities despite pain, which is an important variable tomeasure in interventions designed to enhance self-management. The questionnaire consists of 10 itemsrated on a 7-point scale anchored at 0 with ‘not at allconfident’ and 6 with ‘completely confident’. Items aresummed to generate a total possible score of 60.

Satisfaction: satisfaction with life, global improvement,treatment credibilitySatisfaction with treatment will be assessed by patientsrating their overall improvement in terms of PatientGlobal Impression of Change (PGIC), their satisfactionwith outcome and how credible they found their treat-ment.47 This has five items scored on 11-point scalesranging from 0 (not at all) to 10 (completely). Life satis-faction will be assessed by a single item: ‘All things con-sidered, how satisfied are you with your life as a wholenowadays?’ Responses are on a scale from 0 (extremelydissatisfied) to 10 (extremely satisfied).

Health economics: EQ-5D-5L and SF-6DThis study is an important opportunity to pilot methodsfor estimating the economic impact of interventions onCLBP needed to design cost-effectiveness analyses(CEAs) in future definitive trials. Previously reportedCEAs in the UK have relied on utility values derivedfrom two different instruments—the EQ-5D-5L36 52 53

and the SF-6D.18 The EQ-5D-5L is the most commonlyused tool for generating QALYs; however, the SF-6D,which is derived from scores obtained on the MedicalOutcomes Survey Short Form-12 (v2), may be more sen-sitive to change in CLBP. The economic burden ofCLBP is considerable from an NHS and patient perspec-tive. A resource use questionnaire which identifies keycost drivers (NHS and non-NHS) will be developedbased on previous studies. This will then be piloted toensure completion rates, avoid redundant questions andto identify any additional resource use items sensitive tochange in a CLBP population.54 This pilot study willcompare the validity and sensitivity of the EQ-5D-5L(the most recent version of the EQ-5D) and the SF-6Dfor use in economic evaluations of cognitively enhancedphysiotherapy for CLBP.

Adherence to PACT treatmentPatients’ adherence to PACT treatment will be recordedby trial physiotherapists on a database. Attending bothface-to-face sessions will be considered adherence toPACT treatment. In the UC arm, the type of treatmentand attendance at physiotherapy sessions will berecorded by the trial physiotherapists on the database.Any modifications or departures from randomised treat-ments, withdrawal of participants from trial treatment orresearch follow-up will be recorded and reported assuch.

Qualitative componentPatient interviewsA nested qualitative study will explore patients’ experi-ences of PACT treatment. The aim of these methods willbe to assess patients’ views of the acceptability of PACT,to provide insight into the quantitative results and toexplore processes of change. Semistructured face-to-faceinterviews will be conducted with up to 25 participants(sampled purposively to encompass a mix of gender,age, recruitment site and baseline RMDQ scores) follow-ing their 3-month follow-up assessment (RCT outcomeprimary end point). Interviews will be transcribed verba-tim and analysed using thematic analysis55 to generatethe key themes. Analysis will commence after the firstinterview in an iterative process, allowing early insightsto be explored more fully in later interviews and topicguides to be amended as necessary. A reflexive diary willbe kept during the recruitment, interview and data ana-lysis process to ensure transparency of the analysisprocess. Respondent validity and independent coding byanother researcher will be conducted to check the validityof emergent themes.

8 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

Physiotherapist interviewsAdditional, nested, longitudinal, qualitative methods willexplore the feasibility and acceptability of the PACTtraining programme for physiotherapists. All phy-siotherapists trained in PACT will be invited to attendindividual face-to-face semistructured interviews by inde-pendent researchers. Later, the eight physiotherapistsproviding PACT treatment will be interviewed on twomore occasions, 6 months after training and at the endof treatment delivery, to assess their perceptions of deli-vering this novel physiotherapy service treatment. Allqualitative interviews will provide insight into the accept-ability and feasibility of PACT, development of compe-tency and any contextual factors linked to delivery toinform in any future research in this area.Interviews will be conducted by independent research-

ers, transcribed verbatim and analysed using frameworkanalysis to generate the key themes.56 Analysis will com-mence after the first interview in an iterative process,allowing early insights to be explored more fully in laterinterviews and topic guides to be amended as necessary.A reflexive diary will be kept during the recruitment,interview and data analysis process to ensure transpar-ency of the analysis process. Respondent validity andindependent coding by two researchers will be con-ducted to check the validity of emergent themes.

Proposed sample sizeThe sample size of 240 will provide adequate power todetect a standardised mean difference of 0.40 in theprimary outcome (RMDQ; 5% significance, 80% power)assuming attrition of 20%. Using data from Critchleyet al (2007) and our own initial feasibility study,57 30 thisequates to a 2-point difference between groups, where a2–3 point difference in the RMDQ score is consideredclinically important.51 It is hoped attrition will be mini-mised by a full explanation prior to recruitment of thetime commitment required and importance of complet-ing all follow-up questionnaires. A protocol will be devel-oped to ensure an optimum and standardised follow-upprocess, including recording multiple contact addresses,email addresses and phone numbers.

Statistical analysisThe statistical analysis plan has been approved by theTrial Steering Committee (TSC). The trial will deter-mine the efficacy of the PACT intervention within sixsecondary care physiotherapy clinics. The main efficacyanalysis will be performed only once the database hasbeen cleaned and locked.Stata 12.1 or higher will be used for the descriptive

and main inferential analyses. The main efficacy analysisfor primary, secondary and process outcomes will followan intention-to-treat framework, whereby participantsare analysed according to the groups to which they wererandomised. The analysis will be conducted by the trialstatistician (SN) blind to group allocation. SN will onlybe unblinded once the main efficacy analysis has been

completed. Between-group differences (treatment effi-cacy) will be estimated for the primary outcome RMDQat the postintervention 3-month and 12-month follow-upassessments using linear mixed-effects models. Randomeffects for the intercept and time will be included in themodel. Treatment group, time and a treatment by timeinteraction term will be included as covariates to allowestimates of treatment effect at each time point to becalculated. In addition, a random effect for physiother-apist will be included to account for the partial cluster-ing within physiotherapists in the intervention arm.Estimation of the treatment effects on the secondaryand process outcomes will employ the same method asthe primary efficacy analysis. All outcome variables arecontinuous.Planned secondary analysis will be performed to deter-

mine whether the treatment effect occurs via changes inthe process variables as hypothesised (pain acceptanceand committed action). Specifically, the proportion ofthe treatment effect for disability (RMDQ), QoL (WSAS)and mood (PHQ-9 and GAD-7) at each follow-up that ismediated by the treatment effect on the process variablesat 3 months will be determined. This will be estimated bythe product of coefficients method using bootstrappedSEs.57 This analysis will be undertaken irrespective ofachieving the statistical significance. Where the treat-ment effect is non-significant, additional further analysiswill be conducted to determine the role of post-randomisation effect modifiers in the negative result(adherence, treatment fidelity and therapeutic alli-ance). For example, should there be considerable non-adherence to the treatment, the efficacy of the treatmentfor those who adhere to treatment will also be estimatedin terms of the complier average causal effect (CACE).

ETHICS AND DISSEMINATIONEthical issuesThe trial will be conducted in accordance with currentguidelines for ethical research conduct and subject tofull Research Ethics Committee (REC) approval(National Research Ethics Committee South Central—Berkshire; 14/SC/0277), including any provisions ofsite-specific assessment, and local research and develop-ment approval. It will comply with the InternationalConference for Harmonisation of Good Clinical Practice(ICH GCP) guidelines and the Research GovernanceFramework for Health and Social Care. The trial is regis-tered on a trial registry (ISRCTN95392287) and the leadsite (Guy’s & St Thomas’ NHS Foundation Trust; GSTT)will audit this project annually to ensure compliancewith the necessary legislation.All patients in the trial will benefit from receiving

physiotherapy for their CLBP. This is a very low-risk studyas both treatments are non-invasive and delivered byappropriately qualified physiotherapists. Patients attend-ing PACT sessions should visit hospital less often and willreceive additional resources to aid self-management of

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 9

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

their condition, possibly reducing its impact on partici-pants’ lives. This may lead to benefits for society and theNHS, such as reduced healthcare usage and less time offwork, which is important in such a widespread and costlycondition. The disadvantage of taking part is the add-itional time spent completing the questionnaires and forsome patients an interview. These should not take morethan 60 min in total to complete. Potential participantswill be fully informed of the trial procedures beforeentering the study via a patient information sheet.

Informed consentPotential participants will be identified by clinical phy-siotherapists from each hospital centre, informed aboutthe RCT in writing and invited to participate. Thephysiotherapist will explain that participation is com-pletely voluntary and that they are free to refuse involve-ment. They will be given at least 24 hours to considerwhether they would like to participate. The RAs willthen contact them to see if they are interested in partici-pating and answer any questions about the study, priorto conducting the screening process and signing theconsent form.

Fair accessAny adult patient referred to physiotherapy with lowback pain lasting over 12 weeks and good English will beeligible for the trial. Participants will be able to com-plete measures online, by post or in person, so theyshould not be disadvantaged if they do not have accessto the internet.

DisseminationThe results of this study will be communicated to parti-cipants at the end of the study and disseminated viapeer-reviewed publications, patient interest groups andconference presentations. The results will enable clini-cians, patients and health service managers to makeinformed decisions regarding the efficacy of PACT forpatients with CLBP. However, further studies will benecessary to demonstrate the generalisability of thefindings beyond physiotherapy services in Londonand the South East, as well its effectiveness andcost-effectiveness.

Service user involvementCLBP patients have been involved in the design of thePACT trial and service users have contributed tothe development of the patient guide. Participants fromthe feasibility study have also provided input and feed-back on the proposals for this RCT. One of them is nowthe patient and public involvement (PPI) representativefor this study, providing ongoing input (informal feed-back and participating in TSC meetings) to ensure itaddresses issues relevant to users.

Research governanceThis study will be conducted in accordance with theICH GCP guidelines and the Research GovernanceFramework for Health and Social Care. King’s CollegeLondon is the sponsor of the RCT.The TSC will meet every 6 months to oversee the trial

procedures and ensure good conduct of the study. TheTSC has an independent chair and two independentmembers plus a PPI representative. The trial manage-ment team (EG, DC, LM and the RAs) will holdmonthly meetings to ensure the smooth running of thetrial. The RAs will circulate a monthly newsletter to sta-keholders to review progress relative to the project planand highlight any issues that need to be addressed.Members of the team will consult each other immedi-ately by email and/or phone about any issues that arisebetween meetings.

Monitoring and auditingThe study will be monitored and audited in accordancewith King’s College London procedures. All trial-relateddocuments will be made available on request for moni-toring and audit by King’s College London, trial NHSpartners, the REC and other licencing bodies.

Assessment of safetyAll patients will be assessed and treated by an experi-enced grade 6 or 7 physiotherapist.

Adverse eventsAny adverse events will be recorded by the treatingphysiotherapist in the clinical notes and reported to theRA and chief investigator immediately via email. Patientswill also be offered the opportunity to report anyadverse events on the follow-up questionnaires. If apatient becomes distressed during treatment, then thePACT physiotherapists will be adequately trained to dealwith this or to identify a need for more input/supportand refer them for an appropriate assessment. Therewill be clinical supervisors available at each research siteif needed, and experienced psychologists will supervisephysiotherapists delivering PACT on a monthly basis.

Serious adverse eventsAn adverse event is defined as serious if it results in anoutcome which is life-changing/threatening, disablingor incapacitating. Any serious adverse events that arerecorded will be immediately referred to the chief inves-tigator, who will assess whether it is an adverse reactionthat is classed as serious, whether it could have beencaused by the intervention and whether it is unex-pected. Any serious adverse reactions will be reported tothe TSC for monitoring and advice. They will advisewhether the participant should be withdrawn fromeither their randomised treatment or from the trial.Arrangements will be made by the trial team for furtherassessment and management as agreed with the relevant

10 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

authorities, GP and participant. A report of the outcomewill be provided to the TSC within 1 month.

Stopping rulesThe trial may be stopped prematurely by the sponsor orchief investigator on the basis of new safety informationor for other reasons given by the TSC, regulatory author-ity or ethics committee concerned. The trial may behalted on the advice of the TSC if recruitment rates aresubstantially below expected levels with no possibility ofremedial action or if there are serious adverse reactionsattributable to the trial which mean it is unsafe to con-tinue. If the study is terminated prematurely, active parti-cipants will be informed and no further participant datawill be collected.

Data storageData will be collected and retained in accordance withthe Data Protection Act 1998. The Data ProtectionPolicy of King’s College London will be complied with.The responses to questionnaires will be stored in ananonymised form on a password-protected universitycomputer. The anonymised paper questionnaires will bestored in a locked filing cabinet at Guy’s Campus, King’sCollege London. Study documents (paper and elec-tronic) will be retained in a secure location during andafter the trial has finished. All source documents will beretained for a period of 5 years following the end of thestudy. The chief investigator will be the custodian of thedata and the data will only be used by the study team.

ConclusionsThis paper describes the protocol for the PACT study.The PACT study RCT will assess the feasibility andacceptability of delivering a novel psychologicallyinformed physiotherapy intervention with staff andpatients. It is the first trial to test the efficacy of anACT-informed, physiotherapist-delivered intervention forCLBP. It is noted that further studies will be necessary todemonstrate the generalisability of the findings beyondphysiotherapy services in London and the South East, aswell its effectiveness and cost-effectiveness.

Author affiliations1Department of Psychology, Institute of Psychiatry, Psychology andNeuroscience, King’s College London, London, UK2Faculty of Life Sciences and Medicine, Division of Health and Social CareResearch, Department of Physiotherapy, King’s College London, London, UK3Pain Management Research Institute, Sydney Medical School—Northern,Royal North Shore Hospital, Sydney, New South Wales, Australia4King’s College Hospital NHS Foundation Trust, London, UK5Faculty of Life Sciences and Medicine, Division of Health and Social CareResearch, Department of Physiotherapy, King’s College London, London, UK

Contributors EG, LM, SN, RM-M, DS, MB, JP and DC were involved in thedesign of the study and were coapplicants for funding. EG wrote the first draftof the grant application and is chief investigator. LM, DC and EG developedthe PACT physiotherapy protocols and training included in the trial. RM-Mcontributed to the design and management of the study. LM, DC and EG aresupervising the clinical physiotherapists delivering PACT. DC leads on thehealth economic evaluation. SN leads on statistical analysis. MGH and VW are

leading on the nested qualitative studies and are responsible for trialmanagement. JP is an expert patient advisor, and MB is leading clinicalrecruitment and management at KCH. DS helped develop the PACT treatmentpackage and delivered treatment in the feasibility study. All authors read andapproved the final manuscript.

Funding This article presents independent research funded by the NationalInstitute for Health Research (NIHR) under its Research for Patient Benefit(RfPB) Programme (Grant Reference Number PB-PG-1112-29055). The viewsexpressed are those of the authors and not necessarily those of the Sponsor,the NHS, the NIHR or the Department of Health.

Competing interests None declared.

Patient consent Obtained.

Ethics approval National Research Ethics Committee South Central—Berkshire; 14/SC/0277.

Provenance and peer review Not commissioned; externally peer reviewed.

Data sharing statement Data will be collected and retained in accordancewith the Data Protection Act 1998. The Data Protection Policy of King’sCollege London will be complied with. The responses to questionnaires willbe stored in an anonymised form on a password-protected universitycomputer. The anonymised paper questionnaires will be stored in a lockedfiling cabinet at Guy’s Campus, King’s College London. Study documents(paper and electronic) will be retained in a secure location during and afterthe trial has finished. All source documents will be retained for a period of5 years following the end of the study. The chief investigator will be thecustodian of the data and the data will only be used by the study team.

Open Access This is an Open Access article distributed in accordance withthe terms of the Creative Commons Attribution (CC BY 4.0) license, whichpermits others to distribute, remix, adapt and build upon this work, forcommercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/

REFERENCES1. Hoy D, March L, Brooks P, et al. The global burden of low back pain:

estimates from the Global Burden of Disease 2010 study. AnnRheum Dis 2014;73:968–74.

2. Murray CJL, Vos T, Lozano R, et al. Disability-adjusted life years(DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: asystematic analysis for the Global Burden of Disease Study 2010.Lancet 2012;380:2197–223.

3. Dagenais S, Caro J, Haldeman S. A systematic review of low backpain cost of illness studies in the United States and internationally.Spine J 2008;8:8–20.

4. Department of Health. Departmental Report 2009. London: DoH,2009.

5. Maniadakis N, Gray A. The economic burden of back pain in the UK.Pain 2000;84:95–103.

6. National Institute for Health and Clinical Excellence. Low back pain:early management or persistent non-specific low back pain. CG88.London: National Institute for Health and Clinical Excellence, 2009.http://guidance.nice.org.uk/CG88/Guidance/pdf/English

7. Brox JI, Storheim K, Grotle M, et al. Systematic review of backschools, brief education, and fear-avoidance training for chronic lowback pain. Spine J 2008;8:948–58.

8. Vibe Fersum K, O’Sullivan P, Skouen JS, et al. Efficacy ofclassification-based cognitive functional therapy in patients withnon-specific chronic low back pain: a randomized controlled trial.Eur J Pain 2013;17:916–28.

9. Artus M, van der Windt DA, Jordan KP, et al. Low back painsymptoms show a similar pattern of improvement following a widerange of primary care treatments: a systematic review of randomizedclinical trials. Rheumatology (Oxford) 2010;49:2346–56.

10. van Middelkoop M, Rubinstein SM, Verhagen AP, et al. Exercisetherapy for chronic nonspecific low-back pain. Best Pract Res ClinRheumatol 2010;24:193–204.

11. Foster NE, Delitto A. Embedding psychosocial perspectives withinclinical management of low back pain: integration of psychosociallyinformed management principles into physical therapist practice—challenges and opportunities. Phys Ther 2011;91:790–803.

Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548 11

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from

12. Gatchel RJ, Peng YB, Peters ML, et al. The biopsychosocialapproach to chronic pain: scientific advances and future directions.Psychol Bull 2007;133:581.

13. Eccleston C, Williams A, Morley S. Psychological therapies for themanagement of chronic pain (excluding headache) in adults.Cochrane Database Syst Rev 2009;2:CD007407.

14. Hoffman BM, Papas RK, Chatkoff DK, et al. Meta-analysis ofpsychological interventions for chronic low back pain. Health Psychol2007;26:1–9.

15. Scascighini L, Toma V, Dober-Spielmann S, et al. Multidisciplinarytreatment for chronic pain: a systematic review of interventions andoutcomes. Rheumatology (Oxford) 2008;47:670–8.

16. Williams AC, Eccleston C, Morley S. Psychological therapies for themanagement of chronic pain (excluding headache) in adults.Cochrane Database Syst Rev 2012;11:CD007407.

17. Donaghy M, Nicol M, Davidson KM. Cognitive behaviouralinterventions in physiotherapy and occupational therapy. Edinburgh:Elsevier/Butterworth-Heinemann, 2008.

18. Lamb SE, Hansen Z, Lall R, et al., Back Skills Training Trialinvestigators. Group cognitive behavioural treatment for low-backpain in primary care: a randomised controlled trial andcost-effectiveness analysis. Lancet 2010;375:916–23.

19. Brunner E, De Herdt A, Minguet P, et al. Can cognitive behaviouraltherapy based strategies be integrated into physiotherapy for theprevention of chronic low back pain? A systematic review. DisabilRehabil 2013;35:1–10.

20. Alexanders J, Anderson A, Henderson S. Musculoskeletalphysiotherapists’ use of psychological interventions: a systematicreview of therapists’ perceptions and practice. Physiotherapy2015;101:95–102.

21. Hayes SC, Strosahl K, Wilson KG. Acceptance and commitmenttherapy: an experimental approach to behaviour change. New York:Guilford Press, 1999.

22. McCracken L, Morley S. The psychological flexibility model: a basisfor integration and progress in psychological approaches to chronicpain management. J Pain 2014;15(3):221–34.

23. Vowles KE, Thompson M. Acceptance and commitment therapy forchronic pain, mindfulness and acceptance in behavioral medicine:current theory and practice. Oakland, CA: New Harbinger2011:31–60.

24. Hann KEJ, McCracken LM. A systematic review of randomizedcontrolled trials of acceptance and commitment therapy for adultswith chronic pain: outcome domains, design quality, and efficacy.J Context Behav Sci 2014;3:217–27.

25. Veehof MM, Oskam MJ, Schreurs KM, et al. Acceptance-basedinterventions for the treatment of chronic pain: a systematic reviewand meta-analysis. Pain 2011;152:533–42.

26. Öst L-G. The efficacy of Acceptance and Commitment Therapy: anupdated systematic review and meta-analysis. Behav Res Ther2014;61:105–21.

27. Vowles KE, McCracken LM, O’Brien JZ. Acceptance andvalues-based action in chronic pain: a three-year follow-up analysisof treatment effectiveness and process. Behav Res Ther2011;49:748–55.

28. Pincus T, Anwar S, McCracken LM, et al. Delivering an OptimisedBehavioural Intervention (OBI) to people with low back pain withhigh psychological risk; results and lessons learnt from afeasibility randomized controlled trial of Contextual CognitiveBehavioural Therapy (CCBT) vs. BMC Musculoskelet Disord2015;16:147.

29. Barker E, McCracken LM. From traditional cognitive–behaviouraltherapy to acceptance and commitment therapy for chronic pain: amixed-methods study of staff experiences of change. Br J Pain2014;8:98–106.

30. Critchley DJ, McCracken L, Talewar R, et al. Physiotherapy informedby acceptance and commitment therapy for persistent low back pain:the PACT study. Physiotherapy 2015;101:E277.

31. Frost H, Lamb S, Doll H, et al. Randomised controlled trial ofphysiotherapy compared with advice for low back pain. BMJ2004;329:708.

32. Hay E, Mullins R, Lewis M, et al. Comparison of physical treatmentsversus a brief pain-management programme for back pain in primarycare: a randomised clinical trial in physiotherapy practice. Lancet2005;365:2024–30.

33. Plumb JC, Vilardaga R. Assessing treatment integrity in Acceptanceand Commitment Therapy: strategies and suggestion. Int J BehavConsultation Ther 2010;6:264–95.

34. Godfrey E, Chalder T, Ridsdale L, et al. Investigating the activeingredients of cognitive behaviour therapy and counselling forpatients with chronic fatigue in primary care: developing a newprocess measure to assess treatment fidelity and predict outcome.Br J Clin Psychol 2007;46(Pt 3):253–72.

35. Boutron I, Moher D, Altman DG, et al. Extending the CONSORTstatement to randomized trials of nonpharmacologic treatment:explanation and elaboration. Ann Intern Med 2008;148:295–309.

36. Herdman M, Gudex C, Lloyd A, et al. Development and preliminarytesting of the new five-level version of EQ-5D (EQ-5D-5L). Qual LifeRes 2011;20:1727–36.

37. Ware J, Kosinski M, Keller S. A 12-Item Short-Form Health Survey:construction of scales and preliminary tests of reliability and validity.Medical Care 1996;34:220–3.

38. Roland M, Morris R. A study of the natural history of back pain. PartI: development of a reliable and sensitive measure of disability inlow-back pain. Spine 1983;8:141–4.

39. Fish RA, McGuire B, Hogan M, et al. Research papers: validation ofthe Chronic Pain Acceptance Questionnaire (CPAQ) in an internetsample and development and preliminary validation of the CPAQ-8.Pain 2010;149:435–43.

40. McCracken LM, Chilcot J, Norton S. Further development in theassessment of psychological flexibility: a shortened CommittedAction Questionnaire (CAQ-8). Eur J Pain 2015;19:677–85.

41. Stratford P, Gill C, Westaway M, et al. Assessing disability andchange on individual patients: a report of a patient specific measure.Physiotherapy Canada 1995;47:258–63.

42. Mundt JC, Marks IM, Shear MK, et al. The work and socialadjustment scale: a simple measure of impairment in functioning. BrJ Psychiatry 2002;180:461–4.

43. Nicholas MK. The pain self-efficacy questionnaire: taking pain intoaccount. Eur J Pain 2007;11:153–63.

44. Spitzer RL, Kroenke K, Williams JB, et al. A brief measure forassessing generalized anxiety disorder: the GAD-7. Arch Intern Med2006;166:1092–7.

45. Kroenke K, Spitzer RL, Williams JBW. The PHQ-9: validity of a briefdepression severity measure. J Gen Intern Med 2001;16:606–13.

46. Bombardier C. Outcome assessments in the evaluation of treatmentof spinal disorders—Summary and general recommendations. Spine(Phila Pa 1976) 2000;25:3100–3.

47. Borkovec TD, Nau SD. Credibility of analogue therapy rationales.J Behav Ther Exp Psychiatry 1972;3:257–60.

48. Lamb SE, Lall R, Hansen Z, et al. A multicentred randomisedcontrolled trial of a primary care-based cognitive behaviouralprogramme for low back pain. The Back Skills Training (BeST) trial.Health Technol Assess 2010;14:1–iv.

49. Dworkin RH, Turk DC, Farrar JT, et al. Core outcome measures for chronicpain clinical trials: IMMPACT recommendations. Pain 2005;113:9–19.

50. Roland M, Fairbank J. The Roland-Morris Disability Questionnaireand the Oswestry Disability Questionnaire. Spine 2000;25:3115–24.

51. National Institute for Health and Clinical Excellence. Guide to themethods of technology appraisal. London: National Institute forHealth and Clinical Excellence, 2013.

52. Manca A, Kumar K, Taylor RS, et al. Quality of life, resourceconsumption and costs of spinal cord stimulation versus conventionalmedical management in neuropathic pain patients with failed backsurgery syndrome (PROCESS trial). Eur J Pain 2008;12:1047–58.

53. Henchoz Y, Pinget C, Wasserfallen J-B, et al. Cost-utility analysis ofa three-month exercise programme vs. routine usual care followingmultidisciplinary rehabilitation for chronic low back pain. J RehabMed 2010;42:846–52.

54. Braun V, Clarke V. Using thematic analysis in psychology. Qual ResPsychol 2006;3:77–101.

55. Ritchie J, Lewis JE. Qualitative research practice: a guide for socialscience students and researchers. London: SAGE, 2003.

56. Critchley DJ, Ratcliffe J, Noonan S, et al. Effectiveness andcost-effectiveness of three types of physiotherapy used to reducechronic low back pain disability: a pragmatic randomized trial witheconomic evaluation. Spine 2007;32:1474–81.

57. MacKinnon DP. Introduction to Statistical Mediation Analysis.New York, NY: Taylor & Francis Group, 2008.

12 Godfrey E, et al. BMJ Open 2016;6:e011548. doi:10.1136/bmjopen-2016-011548

Open Access

on March 3, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-011548 on 7 June 2016. D

ownloaded from